What Would a Land Value Tax Actually Do? (Neidle / Tax Policy Associates, 2026)

Summary

"What would a land value tax actually do?" (12 July 2026) is a long, heavily modelled analysis by Dan Neidle, founder of Tax Policy Associates (TPA), a UK non-profit tax-policy shop Neidle set up after a career as a City tax lawyer (Clifford Chance). TPA is not a Georgist organisation — it is a general tax-policy watchdog known for debunking overstated revenue claims across the political spectrum, including its own side's; its 2024 analysis of the Liberal Democrats' proposed "buyback tax" on share buybacks concluded the policy would raise close to nothing once behavioural response was accounted for, a conclusion the Institute for Fiscal Studies independently endorsed.[2] That reputation for calling weak numbers weak either way is part of why this piece carries weight: Neidle spends most of the article building the case for a UK land value tax, then spends the back half warning, in detail, about exactly how it could be done badly.

Model and Method

Neidle built an interactive public model of English residential property, layering:

• HM Land Registry price-paid data (1995–2026)
• ONS postcode-to-geography lookups and the UK House Price Index
• Valuation Office Agency (VOA) council tax band records

to estimate current home values by local authority and council tax band, then split each property's value into land and building components using the abstraction method (estimated sale value minus estimated rebuild cost of the structure). The land/building split rests on a single calibrating assumption: "land equals 55% of value in the retained sold-home sample" nationally. Neidle is explicit that this number is the model's weak point: "The land/building split cannot be verified. Urban vacant sales are unusual, so the overall level rests on assumed rebuild costs and a 55% land-share calibration. These are assumptions."

The model is explicitly static: it holds behaviour fixed and does not attempt to project construction responses, avoidance, migration, or price adjustment beyond the one-off capitalisation effect described below. It also covers residential property only — no commercial land, agricultural land, or non-residential exemptions are modelled.

Headline Findings

Replacing English residential council tax and stamp duty land tax (current combined yield £56.7bn) with a flat-rate land value tax, Neidle calculates a revenue-neutral rate of 1.28% of land value. Building in transitional relief for recent stamp-duty payers pushes the rate to 1.41%; adding a deferral option for cash-poor owners raises it again to 1.46%.

• Most households would pay less. By Neidle's estimate, "at least 63% of people will benefit in immediate cash terms" — concentrated in lower-value regions of the North and Midlands.
• Worked examples. A four-bedroom Blackpool house currently paying roughly £2,100 in council tax would owe about £459 in LVT. An Islington flat whose council tax is £2,000 would face an LVT bill of about £5,000 — some £3,600 more than the equivalent Band D council tax bill. The highest-value Kensington & Chelsea properties would face an average LVT charge of roughly £53,000 a year.
• Capitalisation is concentrated, not uniform. Neidle argues the aggregate price effect is misleading if read as a single national number: "capitalisation under a full-fat LVT is not a national crash. It's an average nothing, disguising a large, concentrated fall at the top — prime London and the commuter belt." Cheaper regions gain in land-price terms what the most expensive properties lose.
• Administration. Neidle estimates ongoing administrative cost (valuation infrastructure, HMRC/VOA capacity) at "somewhere between £300m and £1bn each year."

Limitations Neidle Himself Flags

The article is unusually candid about its own weaknesses, which is precisely why the wiki treats it as a serious source rather than an advocacy piece:

1. The 55% land-share calibration cannot be independently verified — the abstraction method's core input, since genuinely unimproved urban land rarely trades in a way that lets the split be checked directly. Neidle's sensitivity analysis shows the revenue-neutral rate could plausibly range from roughly 1.1% to 1.7% depending on this single assumption.
2. No behavioural modelling. Construction responses, avoidance, and migration are excluded; the model shows only the mechanical incidence of a like-for-like revenue swap, not the dynamic effects Georgist theory (or its critics) predict over time.
3. Historical implementation failures, cited by Neidle as reasons for caution rather than reasons to abandon the idea: the under-yielding 1909–1914 Lloyd George land duties, the suspension of Denmark's grundskyld valuation base at points in its history, and Welsh Government valuation research, which Neidle reads as showing no assessment methodology is yet accurate enough for live implementation. That last reading overstates the primary sources: the Welsh programme's published papers are desk-based scoping reviews proposing five method families for future field testing (tendered in 2025, results not yet published), not completed accuracy tests — see the dedicated research page for what the programme has actually found so far.
4. No modelling of farmland relief, social housing treatment, or council-tax-support scheme interactions — all of which any real bill would have to address and none of which the static model attempts.

The Verdict

Neidle's own conclusion is a considered, conditional "yes": "The principled case is overwhelming, and the practical prize is real and large." But he pairs that with an explicit warning against doing it carelessly, prescribing a long phase-in (a decade or more), transitional relief for recent buyers, deferral for asset-rich but cash-poor owners, and probably some regional rate variation to blunt the sharpest London losses. His closing framing: "a reform this big, reaching into the home that is most people's largest asset, has to be done slowly or not at all."

Where This Fits the Wiki's Existing Evidence

Several of Neidle's own caveats track objections the wiki already carries in detail, and a reader weighing this piece should read them alongside it rather than in isolation:

• The unverifiable 55% land-share assumption is the UK-specific version of the debate on whether land value can be assessed accurately — that page's "Limits and Caveats" section already concedes that assessment quality is uneven and that at least one real mass-valuation system has badly mistracked market value; Neidle's model shows the same fragility can sit underneath a rigorous, modern, data-rich UK exercise, not just historical or developing-country cases.
• Neidle's capitalisation finding — sharp, concentrated falls at the top rather than a uniform national effect — sits inside the same empirical dispute the wiki documents on LVT's one-time transition wealth shock, where two Danish quasi-experiments (DØRS and Nielsson, Wroblewski & Yding) reach opposite conclusions on how much of a future land tax capitalises into present prices at all.
• His prescribed remedies — phase-in, deferral, transitional relief — are exactly the standard toolkit the wiki's asset-rich, cash-poor objection page describes as the accepted answer to the "little old lady" problem, applied here to a live, numbered UK proposal rather than a generic design discussion.
• His citation of the 1909 Budget and Danish/Welsh valuation difficulties as cautionary precedent is the same historical record examined, from the opposite rhetorical direction, in the Watling–Doucet exchange summarised on this wiki's companion research page — Neidle reads the same 1909 failure as "be careful," Watling reads it as "it can't be done," and Doucet replies that contemporaneous successes (Vancouver, Houston, 1920s New York, Qingdao) undercut the generalisation.

Limits of This Source

TPA is a policy-advocacy and journalism-adjacent think tank, not a peer-reviewed academic source (source-quality tier 4–5 under the wiki's own hierarchy), and the piece is a single-author model without independent replication. Its land/building split is, by the author's own admission, unverifiable with current UK data. The wiki cites it for what it is: an unusually transparent and self-critical worked example of what a specific, numbered UK LVT reform would do to specific households — not as a peer-reviewed incidence estimate.

See Also

• Objection: Land value can't be assessed accurately — the general version of the land-share-calibration problem this model exposes concretely
• Objection: LVT inflicts a one-time wealth shock on current owners — the capitalisation and phase-in questions Neidle's regional winners/losers map illustrates
• Objection: LVT hurts the "asset-rich, cash-poor" — the deferral and transitional-relief remedies Neidle independently arrives at
• Contra Watling on "The Failure of the Land Value Tax" — the parallel debate over the same historical implementation failures Neidle cites as caution
• United Kingdom — the 1909 People's Budget and the wider UK land-tax record this proposal sits within
• Institute for Fiscal Studies — the mainstream UK public-finance institute whose independent buyback-tax conclusion corroborated TPA's 2024 work
• The Dynamic of a Tax on Land Value (Spring-Ragain, 2025) — a contrasting theoretical/dynamic spatial model of LVT, versus this page's empirical/static microsimulation of a real jurisdiction

Sources

1. Dan Neidle (2026), "What would a land value tax actually do?", Tax Policy Associates, 12 July
2. taxpolicy.org.uk — used for all model methodology, figures, worked examples, and quotations above; fetched and verified verbatim this session for every direct quotation.
3. Dan Neidle (2024), "The Lib Dem buyback tax won't raise £2bn, and could raise nothing," Tax Policy Associates, 9 June 2024. taxpolicy.org.uk — used only for the one-sentence characterisation of TPA's track record of debunking overstated revenue claims and the Institute for Fiscal Studies's independent corroboration of that specific conclusion; this earlier piece is unrelated to land value tax and is not itself scanned in depth.